[linux-2.6-block.git] / Documentation / lzo.txt


LZO stream format as understood by Linux's LZO decompressor
===========================================================

Introduction

  This is not a specification. No specification seems to be publicly available
  for the LZO stream format. This document describes what input format the LZO
  decompressor as implemented in the Linux kernel understands. The file subject
  of this analysis is lib/lzo/lzo1x_decompress_safe.c. No analysis was made on
  the compressor nor on any other implementations though it seems likely that
  the format matches the standard one. The purpose of this document is to
  better understand what the code does in order to propose more efficient fixes
  for future bug reports.

Description

  The stream is composed of a series of instructions, operands, and data. The
  instructions consist in a few bits representing an opcode, and bits forming
  the operands for the instruction, whose size and position depend on the
  opcode and on the number of literals copied by previous instruction. The
  operands are used to indicate :

    - a distance when copying data from the dictionary (past output buffer)
    - a length (number of bytes to copy from dictionary)
    - the number of literals to copy, which is retained in variable "state"
      as a piece of information for next instructions.

  Optionally depending on the opcode and operands, extra data may follow. These
  extra data can be a complement for the operand (eg: a length or a distance
  encoded on larger values), or a literal to be copied to the output buffer.

  The first byte of the block follows a different encoding from other bytes, it
  seems to be optimized for literal use only, since there is no dictionary yet
  prior to that byte.

  Lengths are always encoded on a variable size starting with a small number
  of bits in the operand. If the number of bits isn't enough to represent the
  length, up to 255 may be added in increments by consuming more bytes with a
  rate of at most 255 per extra byte (thus the compression ratio cannot exceed
  around 255:1). The variable length encoding using #bits is always the same :

       length = byte & ((1 << #bits) - 1)
       if (!length) {
               length = ((1 << #bits) - 1)
               length += 255*(number of zero bytes)
               length += first-non-zero-byte
       }
       length += constant (generally 2 or 3)

  For references to the dictionary, distances are relative to the output
  pointer. Distances are encoded using very few bits belonging to certain
  ranges, resulting in multiple copy instructions using different encodings.
  Certain encodings involve one extra byte, others involve two extra bytes
  forming a little-endian 16-bit quantity (marked LE16 below).

  After any instruction except the large literal copy, 0, 1, 2 or 3 literals
  are copied before starting the next instruction. The number of literals that
  were copied may change the meaning and behaviour of the next instruction. In
  practice, only one instruction needs to know whether 0, less than 4, or more
  literals were copied. This is the information stored in the <state> variable
  in this implementation. This number of immediate literals to be copied is
  generally encoded in the last two bits of the instruction but may also be
  taken from the last two bits of an extra operand (eg: distance).

  End of stream is declared when a block copy of distance 0 is seen. Only one
  instruction may encode this distance (0001HLLL), it takes one LE16 operand
  for the distance, thus requiring 3 bytes.

  IMPORTANT NOTE : in the code some length checks are missing because certain
  instructions are called under the assumption that a certain number of bytes
  follow because it has already been garanteed before parsing the instructions.
  They just have to "refill" this credit if they consume extra bytes. This is
  an implementation design choice independant on the algorithm or encoding.

Byte sequences

  First byte encoding :

      0..17   : follow regular instruction encoding, see below. It is worth
                noting that codes 16 and 17 will represent a block copy from
                the dictionary which is empty, and that they will always be
                invalid at this place.

      18..21  : copy 0..3 literals
                state = (byte - 17) = 0..3  [ copy <state> literals ]
                skip byte

      22..255 : copy literal string
                length = (byte - 17) = 4..238
                state = 4 [ don't copy extra literals ]
                skip byte

  Instruction encoding :

      0 0 0 0 X X X X  (0..15)
        Depends on the number of literals copied by the last instruction.
        If last instruction did not copy any literal (state == 0), this
        encoding will be a copy of 4 or more literal, and must be interpreted
        like this :

           0 0 0 0 L L L L  (0..15)  : copy long literal string
           length = 3 + (L ?: 15 + (zero_bytes * 255) + non_zero_byte)
           state = 4  (no extra literals are copied)

        If last instruction used to copy between 1 to 3 literals (encoded in
        the instruction's opcode or distance), the instruction is a copy of a
        2-byte block from the dictionary within a 1kB distance. It is worth
        noting that this instruction provides little savings since it uses 2
        bytes to encode a copy of 2 other bytes but it encodes the number of
        following literals for free. It must be interpreted like this :

           0 0 0 0 D D S S  (0..15)  : copy 2 bytes from <= 1kB distance
           length = 2
           state = S (copy S literals after this block)
         Always followed by exactly one byte : H H H H H H H H
           distance = (H << 2) + D + 1

        If last instruction used to copy 4 or more literals (as detected by
        state == 4), the instruction becomes a copy of a 3-byte block from the
        dictionary from a 2..3kB distance, and must be interpreted like this :

           0 0 0 0 D D S S  (0..15)  : copy 3 bytes from 2..3 kB distance
           length = 3
           state = S (copy S literals after this block)
         Always followed by exactly one byte : H H H H H H H H
           distance = (H << 2) + D + 2049

      0 0 0 1 H L L L  (16..31)
           Copy of a block within 16..48kB distance (preferably less than 10B)
           length = 2 + (L ?: 7 + (zero_bytes * 255) + non_zero_byte)
        Always followed by exactly one LE16 :  D D D D D D D D : D D D D D D S S
           distance = 16384 + (H << 14) + D
           state = S (copy S literals after this block)
           End of stream is reached if distance == 16384

      0 0 1 L L L L L  (32..63)
           Copy of small block within 16kB distance (preferably less than 34B)
           length = 2 + (L ?: 31 + (zero_bytes * 255) + non_zero_byte)
        Always followed by exactly one LE16 :  D D D D D D D D : D D D D D D S S
           distance = D + 1
           state = S (copy S literals after this block)

      0 1 L D D D S S  (64..127)
           Copy 3-4 bytes from block within 2kB distance
           state = S (copy S literals after this block)
           length = 3 + L
         Always followed by exactly one byte : H H H H H H H H
           distance = (H << 3) + D + 1

      1 L L D D D S S  (128..255)
           Copy 5-8 bytes from block within 2kB distance
           state = S (copy S literals after this block)
           length = 5 + L
         Always followed by exactly one byte : H H H H H H H H
           distance = (H << 3) + D + 1

Authors

  This document was written by Willy Tarreau <w@1wt.eu> on 2014/07/19 during an
  analysis of the decompression code available in Linux 3.16-rc5. The code is
  tricky, it is possible that this document contains mistakes or that a few
  corner cases were overlooked. In any case, please report any doubt, fix, or
  proposed updates to the author(s) so that the document can be updated.
Commit	Line	Data
d98a0526 WT	1
	2	LZO stream format as understood by Linux's LZO decompressor
	3	===========================================================
	4
	5	Introduction
	6
	7	This is not a specification. No specification seems to be publicly available
	8	for the LZO stream format. This document describes what input format the LZO
	9	decompressor as implemented in the Linux kernel understands. The file subject
	10	of this analysis is lib/lzo/lzo1x_decompress_safe.c. No analysis was made on
	11	the compressor nor on any other implementations though it seems likely that
	12	the format matches the standard one. The purpose of this document is to
	13	better understand what the code does in order to propose more efficient fixes
	14	for future bug reports.
	15
	16	Description
	17
	18	The stream is composed of a series of instructions, operands, and data. The
	19	instructions consist in a few bits representing an opcode, and bits forming
	20	the operands for the instruction, whose size and position depend on the
	21	opcode and on the number of literals copied by previous instruction. The
	22	operands are used to indicate :
	23
	24	- a distance when copying data from the dictionary (past output buffer)
	25	- a length (number of bytes to copy from dictionary)
	26	- the number of literals to copy, which is retained in variable "state"
	27	as a piece of information for next instructions.
	28
	29	Optionally depending on the opcode and operands, extra data may follow. These
	30	extra data can be a complement for the operand (eg: a length or a distance
	31	encoded on larger values), or a literal to be copied to the output buffer.
	32
	33	The first byte of the block follows a different encoding from other bytes, it
	34	seems to be optimized for literal use only, since there is no dictionary yet
	35	prior to that byte.
	36
	37	Lengths are always encoded on a variable size starting with a small number
	38	of bits in the operand. If the number of bits isn't enough to represent the
	39	length, up to 255 may be added in increments by consuming more bytes with a
	40	rate of at most 255 per extra byte (thus the compression ratio cannot exceed
	41	around 255:1). The variable length encoding using #bits is always the same :
	42
	43	length = byte & ((1 << #bits) - 1)
	44	if (!length) {
	45	length = ((1 << #bits) - 1)
	46	length += 255*(number of zero bytes)
	47	length += first-non-zero-byte
	48	}
	49	length += constant (generally 2 or 3)
	50
	51	For references to the dictionary, distances are relative to the output
	52	pointer. Distances are encoded using very few bits belonging to certain
	53	ranges, resulting in multiple copy instructions using different encodings.
	54	Certain encodings involve one extra byte, others involve two extra bytes
	55	forming a little-endian 16-bit quantity (marked LE16 below).
	56
	57	After any instruction except the large literal copy, 0, 1, 2 or 3 literals
	58	are copied before starting the next instruction. The number of literals that
	59	were copied may change the meaning and behaviour of the next instruction. In
	60	practice, only one instruction needs to know whether 0, less than 4, or more
	61	literals were copied. This is the information stored in the <state> variable
	62	in this implementation. This number of immediate literals to be copied is
	63	generally encoded in the last two bits of the instruction but may also be
	64	taken from the last two bits of an extra operand (eg: distance).
65
66	End of stream is declared when a block copy of distance 0 is seen. Only one
67	instruction may encode this distance (0001HLLL), it takes one LE16 operand
68	for the distance, thus requiring 3 bytes.
69
70	IMPORTANT NOTE : in the code some length checks are missing because certain
71	instructions are called under the assumption that a certain number of bytes
72	follow because it has already been garanteed before parsing the instructions.
73	They just have to "refill" this credit if they consume extra bytes. This is
74	an implementation design choice independant on the algorithm or encoding.
75
76	Byte sequences
77
78	First byte encoding :
79
80	0..17 : follow regular instruction encoding, see below. It is worth
81	noting that codes 16 and 17 will represent a block copy from
82	the dictionary which is empty, and that they will always be
83	invalid at this place.
84
85	18..21 : copy 0..3 literals
86	state = (byte - 17) = 0..3 [ copy <state> literals ]
87	skip byte
88
89	22..255 : copy literal string
90	length = (byte - 17) = 4..238
91	state = 4 [ don't copy extra literals ]
92	skip byte
93
94	Instruction encoding :
95
96	0 0 0 0 X X X X (0..15)
97	Depends on the number of literals copied by the last instruction.
98	If last instruction did not copy any literal (state == 0), this
99	encoding will be a copy of 4 or more literal, and must be interpreted
100	like this :
101
102	0 0 0 0 L L L L (0..15) : copy long literal string
103	length = 3 + (L ?: 15 + (zero_bytes * 255) + non_zero_byte)
104	state = 4 (no extra literals are copied)
105
106	If last instruction used to copy between 1 to 3 literals (encoded in
107	the instruction's opcode or distance), the instruction is a copy of a
108	2-byte block from the dictionary within a 1kB distance. It is worth
109	noting that this instruction provides little savings since it uses 2
110	bytes to encode a copy of 2 other bytes but it encodes the number of
111	following literals for free. It must be interpreted like this :
112
113	0 0 0 0 D D S S (0..15) : copy 2 bytes from <= 1kB distance
114	length = 2
115	state = S (copy S literals after this block)
116	Always followed by exactly one byte : H H H H H H H H
117	distance = (H << 2) + D + 1
118
119	If last instruction used to copy 4 or more literals (as detected by
120	state == 4), the instruction becomes a copy of a 3-byte block from the
121	dictionary from a 2..3kB distance, and must be interpreted like this :
122
123	0 0 0 0 D D S S (0..15) : copy 3 bytes from 2..3 kB distance
124	length = 3
125	state = S (copy S literals after this block)
126	Always followed by exactly one byte : H H H H H H H H
127	distance = (H << 2) + D + 2049
128
129	0 0 0 1 H L L L (16..31)
130	Copy of a block within 16..48kB distance (preferably less than 10B)
131	length = 2 + (L ?: 7 + (zero_bytes * 255) + non_zero_byte)
132	Always followed by exactly one LE16 : D D D D D D D D : D D D D D D S S
133	distance = 16384 + (H << 14) + D
134	state = S (copy S literals after this block)
135	End of stream is reached if distance == 16384
136
137	0 0 1 L L L L L (32..63)
138	Copy of small block within 16kB distance (preferably less than 34B)
139	length = 2 + (L ?: 31 + (zero_bytes * 255) + non_zero_byte)
140	Always followed by exactly one LE16 : D D D D D D D D : D D D D D D S S
141	distance = D + 1
142	state = S (copy S literals after this block)
143
144	0 1 L D D D S S (64..127)
145	Copy 3-4 bytes from block within 2kB distance
146	state = S (copy S literals after this block)
147	length = 3 + L
148	Always followed by exactly one byte : H H H H H H H H
149	distance = (H << 3) + D + 1
150
151	1 L L D D D S S (128..255)
152	Copy 5-8 bytes from block within 2kB distance
153	state = S (copy S literals after this block)
154	length = 5 + L
155	Always followed by exactly one byte : H H H H H H H H
156	distance = (H << 3) + D + 1
157
158	Authors
159
160	This document was written by Willy Tarreau <w@1wt.eu> on 2014/07/19 during an
161	analysis of the decompression code available in Linux 3.16-rc5. The code is
162	tricky, it is possible that this document contains mistakes or that a few
163	corner cases were overlooked. In any case, please report any doubt, fix, or
164	proposed updates to the author(s) so that the document can be updated.